Many organizations have mainframe computers installed, yet the senior management of those organizations seem to be sleep-walking into the future without realizing the power of the mainframe, the amount of revenue that results from the use of the mainframe, and the potential the mainframe has to drive the business forward.
Too often, the mainframe is treated like a ghetto area where things carry on as they always have in their strange arcane way, and all the exciting computing goes on in the world of PCs, phones, and the cloud—although even that is a little bit strange!
It’s because company executives can see changes to the look of their Android phones’ screens, and Windows applications can look different, but those execs don’t realize when they’re using a mainframe. And, if there’s no push to develop applications, they don’t see their existing CICS or IMS applications change in any way.
Too often, conversations about mainframes with people who don’t know much about them end up with the mainframer picturing in their mind a sleek modern piece of architecture—perhaps sitting in a glass rack with blue lighting all around—and the finance manager picturing a mainframe he once saw in an old film with lots of tape drives whirring backward and forward as if they all have parity errors, and someone feeding in punched cards into a reader. They’re both using the same word, but both have a completely different pictures in their heads.
And all too often, senior management don’t realize how useful mainframes can be. They don’t realize that over 70 percent of Fortune 500 companies use IBM mainframes. An estimated 87 percent of all credit card transactions are carried out on a mainframe. According to IBM, mainframes handle 68 percent of the world’s production IT workloads, yet they account for only 6 percent of IT costs. OK, I understand that IBM makes mainframes etc. The source of that information can be found at https://www.ibm.com/it-infrastructure/servers/mainframes. It’s also estimated that mainframes handle more than 30 billion business transactions each day. And 92 of the world’s top 100 banks are mainframe users.
Mainframes are also resilient when it comes to disaster recovery. There’s a story about an insurance company located in lower Manhattan, near the East River that had a large data center in its basement. It had a mainframe backed up to another mainframe located in New York. Following Hurricane Sandy blowing through in 2012, the data center flooded, which destroyed all the servers, including the mainframe. However, the mainframe’s workload was restored within 24 hours from the remote backup. The other servers didn’t have similar disaster recovery in place because it was too expensive. Or there was a data center in Japan that was struck by an earthquake strong enough to destroy all of its x86 machines. The mainframe apparently fell on its side but continued to work.
The biggest criticism about mainframes is that they are expensive. And yet anytime anyone has looked at the costs, this has been shown to be untrue. To be fair, your organization does need to be of a certain size for a mainframe to be cost-effective you, but after that, the figures show mainframes are cheaper. One of the first comparisons was carried out by Xephon back in the 1990s. They looked at server costs and people costs and found mainframes were cheaper. The more recent figures from IBM, mentioned above, also show that mainframes are cheaper. Too often, the CFO and his team will simply look at the cost of acquisition. What they don’t think about is the software, networking, and people costs. There comes a point where increasing the number of servers gets more expensive than running the same workload on a mainframe. On a mainframe, you can have just a few people managing hundreds or even thousands of virtual servers. You’d need far more people to look after other platforms, like Linux.
And it’s not generally known among C-level execs that not only can you run Linux on a mainframe, you can have a mainframe that is Linux. Rather than needing mainframe-experienced staff, you can use your Linux experts to work on your new IBM LInuxONE server and get all the advantages of scaling. You also get all the advantages of fault-tolerant mainframe hardware capable of more than 90,000 I/O operations per second and with a mean time between failure (MTBF) measured in decades. And you can do all this in a much smaller footprint, which gives additional savings from a reduced need for power, cooling, networking hardware and the other infrastructure.
You might well find people criticizing mainframes for being too slow to develop new applications. Every time you want something new done, it takes two years for the software to arrive—and by then, you want something else. This may well have been true of application development in the past, but for the last few years sites have been adopting DevOps.
DevOps isn’t a product that can be bought, it’s really about organizational change. It is the practice of Development and IT Operations working together through the entire software life-cycle, from design through the development process to production. This not only requires a change in behavior and culture, but the implementation of processes and the use of a new tool chain to bring it all together. The main goal of a DevOps approach is to develop and deploy innovation faster—to meet the needs of your customers or staff. Using DevOps, organizations can respond to Line of Business (LOB) requirements faster by leveraging agile development. When development is agile, small teams work interactively on tasks in iterative work cycles with their focus on delivering value to customers. A key feature of DevOps is continuous deployment—which means tests are automated and software is immediately deployed to production. IMS and CICS now use DevOps and updates are produced quarterly.
Of course, you’ll hear the leadership team talk about how old their experienced mainframers are getting and how retirement will mean the loss to the organization of all that mainframe expertise. And, they’ll conclude, it makes sense to migrate all the important applications to other platforms that younger people can work on—and so keep the company in business. Of course, the organization could take on youngsters and start training them. Or they could embrace Zowe.
The Open Mainframe Project, a collaborative project managed by the Linux Foundation, announced Zowe on 28 August 2018. Zowe is the first open-source framework for z/OS, and provides solutions for development and operations teams to securely manage, control, script and develop on the mainframe like any other cloud platform. And that’s the key to mainframe-using organizations being successful in the future—those new developers don’t need to have previous mainframe experience! By using Zowe, non-mainframer developers can use the open-source industry-standard tools they are already familiar with to access mainframe resources and services. Zowe is being developed by IBM, Broadcom and Rocket Software among others—so it’s not some flash-in-the-pan idea.
So, wake up sleepy execs and recognize that the mainframe is probably the most powerful tool your company has to keep your organization in business in the future. Don’t let your competitors see the value of their mainframe first and out compete you.